Multiscale modeling and numerical simulations of Lithium ion battery electrodes using real microstructures
Mathematics
Final Report Abstract
The microstructure of lithium-ion battery electrodes determines their performance. Optimizing the electrode microstructure improves the electrode performance. Thus, the long-term goal of this DFG-funded project was to develop an interdisciplinary method for model-based optimization of porous electrodes. The two project partners significantly advanced the understanding of the microstructure influence on lithium ion battery performance. First, new methods were developed to adequately reconstruct and analyze the microstructure of real electrodes using FIB and X-ray tomography [9], and even by combining both tomographic methods in a correlative approach. Moreover, a novel method for the detailed analysis of graphite anodes was developed, called Laser Tomography. Analogously to FIB tomography, the imaging uses scanning electron microscopy (SEM), benefiting from high resolution and excellent imaging contrast mechanisms. Within this project we achieved the worldwide first reconstruction of a graphite anode using laser tomography (conducted in collaboration with the device manufacturer Thermo Fisher Scientific). The presentation of the achieved results at an international conference was rewarded with the Poster Prize. Furthermore, the methods developed could also be used in other fields to investigate porous structures, for example in the field of solid oxide fuel cells and for catalysts. To optimize electrode structures, it is necessary to study a wide range of different electrode structures. Since production of a variety of cathode sheets on current collectors through laboratory experiments would be very expensive with respect to costs and time, models were developed, capable of generating realistic virtual microstructures with predefined characteristics. Basically, three-dimensional cathode microstructure reconstructions were established by FIB tomography, and the individually shaped particles of the active material phase as well as the carbon black were extracted. A representative set of extracted particles was used to create structures with specified characteristics, such as the desired mass content of active material and conductive additive or specific particle size distributions. In combination with performance models, a preselection of promising design concepts can be conducted. To this aim, performance models were developed in this project. Thereby different types of electrochemical performance models were implemented, verified and compared, namely homogenized (Newman-type) and spatially-resolved models. The homogenized models use the microstructural parameters from the characterization as input parameters, while a workflow was established for the spatially resolved models to enable the use of complex, realistic microstructures as computational grid (based on data obtained from 3D reconstruction or virtually generated). The modelling framework together with the reconstruction methodology derived in this project constitutes a very precise tool for determining the rate-limiting processes in a LIB electrode. The microstructure effects on the electrochemical performance are taken into account in the reduced multiscale models, as well as in the partially homogenized 3D model. In this way, the limiting processes of lithium transport are precisely modelled in both the electrolyte and in the active material. Also a possible limiting electronic transport process can be determined with the accurate reconstruction of conductive additives in the electrodes. Overall, we developed methods which can be used for a model-based optimization of lithium-ion battery electrodes. A preliminary implementation of an electrode-optimization tool has been performed in this project for a virtual two-phase cathode. To find an optimal distribution of active particles, a covariance matrix adaptation evolution strategy (CMA-ES) was coupled with the virtual structure generator and the computation of effective transport parameters. To make this tool realistic, the distribution of additives (binder and carbon black) also needs to be included. The goals achieved in this project are an essential starting point for a follow-up project on the systematic optimization of LIB electrodes.
Publications
-
“On the implementation of the eXtended Finite Element Method (XFEM) for interface problems”, Archive of Numerical Software, Vol. 4(2), pp. 1-23, 2016
T. Carraro, S. Wetterauer
-
Microstructural Characterisation, Modelling and Simulation of Solid Oxide Fuel Cell Cathodes, PhD thesis, KIT Scientific Publishing, Karlsruhe 2017
J. Joos
-
„A goal-oriented dual weighted error estimator for a class of homogenization problems“, Journal of Scientific Computing, 71(3), pp. 1169-1196, 2017
T. Carraro, C. Goll
-
Advanced impedance modelling of Ni/8YSZ cermet anodes, Electrochim Acta, 265, pp. 736-750, 2018
S. Dierickx, J. Joos, A. Weber, and E. Ivers-Tiffée
-
Directional freezecast hybrid-backbone meso-macroporous bodies as micromonolith catalysts for gas-to-liquid processes, Journal of Materials Chemistry A, 1, pp. 1-12, 2018
J. Kim, V. Nese, J. Joos, K. Jeske, N. Duyckaerts, N. Pfänder, and G. Prieto
-
Einfluss von Mikrostruktur und Materialparametern auf die Leistungsfähigkeit poröser Elektroden für Lithium-Ionen Batterien, PhD thesis, KIT Scientific Publishing, Karlsruhe 2018
J. Costard
-
“Effective pressure boundary condition for the filtration through porous medium via homogenization”, Nonlinear Analysis: Real World Applications, 44, pp. 149–172, 2018
T. Carraro, E. Marušić-Paloka, A. Mikelić
-
Microstructural feature analysis of commercial Li-ion battery cathodes by focused ion beam tomography, J. Power Sources, 427, pp. 1-14, 2019
L. Almar, J. Joos, and E. Ivers-Tiffée
